1. Field
Example embodiments of the present disclosure relate to a method that may effectively store image information so that a three-dimensional (3D) video system may generate a plurality of view images with a relatively small amount of data, and more particularly, to an image processing apparatus and method that may obtain and store occlusion region information according to a difference between views.
2. Description of the Related Art
Interest in three-dimensional (3D) imaging is increasing. 3D images may be configured by providing a plurality of images corresponding to different viewpoints. For example, the 3D image may be a multi-view image corresponding to a plurality of views or a stereoscopic image providing a left eye image and a right eye image corresponding to two different views.
In the case of the multi-view image, a color image corresponding to each view may need to be provided. However, to substantially acquire and encode total color images of each view, an amount of data to be stored and be transmitted may be great.
Accordingly, when a color image is appropriately generated by substantially acquiring and storing color images and depth images with respect to a first portion of color images of views associated with the stereoscopic image or the multi-view image, and by using the substantially acquired and stored view images with respect to another portion of the color images of the views, it is possible to significantly decrease an amount of data to be transmitted and be stored.
There is a desire for an image processing method that may appropriately acquire and store occlusion region information dis-occluded by warping occurring due to a view difference.